Qwen3 Coder 480B A35B API Prices Rise: Input +27%, Output +55%

What Changed

Qwen has increased API prices on Qwen3 Coder 480B A35B. Input token pricing moves from $0.30 to $0.38 per 1M tokens, while output pricing jumps from $1.00 to $1.55 per 1M tokens — the headline figure is a +55% increase on the output side.

Why Output Pricing Is What Matters

For most coding workloads, output tokens drive the bill. Code generation, explanation, and refactoring tasks are output-heavy by nature. At $1.55/M output tokens, teams running high-volume agentic pipelines or continuous code review workflows will feel this more than occasional users.

To put it in rough terms: a workload generating 10M output tokens per month previously cost $10.00 in output charges alone — it now costs $15.50. That's a meaningful line item at scale, though still competitive with larger proprietary coding models.

Who Should Care

• High-volume users: The +55% output increase adds up fast. Audit your token usage before dismissing it.
• Low-volume or prototyping users: At these absolute price points, the delta is unlikely to change your tooling decisions.
• Teams evaluating alternatives: If this nudges the cost-quality tradeoff unfavorably for your use case, it's worth comparing against other large coding models currently available.

Check Qwen3 Coder 480B A35B — live specs & price history for up-to-date pricing and benchmark comparisons against competing models on the site.

Bottom Line

The input increase is modest. The output increase is not. If Qwen3 Coder is already in your stack, run the numbers against your actual output token volume before deciding whether to absorb the cost or shop around.

Ezra, Scout AI Team